Reclassifying a raster by defined interval with QGIS

Question

I would like to reclassify a DEM raster into bins of 100 metres. Both the GRASS r.reclass and built-in QGIS tools require predefined classes. I would like to have them produced automatically from the data range of the raster. My current solution is via Excel used to create the reclass list for r.reclass. However, I would like to avoid this step.

Any ideas for an easier way?

Answer 1

You can use python to create the classes to use in reclassifybytable:

import math
rlayer = QgsProject.instance().mapLayersByName('nh_67_4')[0]
stats = rlayer.dataProvider().bandStatistics(1, QgsRasterBandStats.All)

MIN, MAX = stats.minimumValue, stats.maximumValue #142.04449462890625   920.567626953125
n = 100
bins = [math.floor(r/100)*100 for r in range(int(MIN), int(MAX)+n, n)] #[100, 200, 300, 400, 500, 600, 700, 800, 900]
bins+=[bins[-1]+n] #[100, 200, 300, 400, 500, 600, 700, 800, 900, 1000]
bins = [b for b in zip(bins, bins[1:], bins)] #[(100, 200, 100), (200, 300, 200), (300, 400, 300), (400, 500, 400), (500, 600, 500), (600, 700, 600), (700, 800, 700), (800, 900, 800), (900, 1000, 900)]
bins = [item for sublist in bins for item in sublist] #[100, 200, 100, 200, 300, 200, 300, 400, 300, 400, 500, 400, 500, 600, 500, 600, 700, 600, 700, 800, 700, 800, 900, 800, 900, 1000, 900]

processing.runAndLoadResults("native:reclassifybytable",  {'INPUT_RASTER':rlayer,'RASTER_BAND':1,
    'TABLE':bins,'NO_DATA':-9999,'RANGE_BOUNDARIES':0,
    'NODATA_FOR_MISSING':False,'DATA_TYPE':5,'OUTPUT':'TEMPORARY_OUTPUT'})

Answer 2

Use Round Raster with a number of decimal places of -2 to round to multiples of 100:

negative number of decimal places may be used to round values to powers of a base n. QGIS help

Use round up/down options for different types of classes (e.g. 300 to 400, 400 to 500 instead of 350 to 450, 450 to 550 etc.).

Left: original, right: reclassified result: